Publication

Abstract : Processing sequence of data of humongous length that may or may not vary over time in an efficient manner is the need of the hour with large amount of data available without inferences being made. Massive online analytics and Stream Analytics can be very helpful in this case. Healthcare data are more critical and may need immediate attention. One such challenge for healthcare related analytics is Diabetes which is a largely growing non-communicable disease. Prediction of diabetes can help in early intervention to control the disease. Stream Analytics is suitable for this area very much. The proposed work performs classification using massive online analytics and improvises the results of classification based on Hoeffding tree optimization using ozaboost techniques.